TC Electronics - Stereo Chorus Flanger [pcb+layout]  [schematic]

All about modern commercial stompbox circuits from Electro Harmonix over MXR, Boss and Ibanez into the nineties.
User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

First you must know how BBD works, then you must know what Vgg affects, then you must know how bias voltage and Vgg interact each other, then you must know what's BBD's signal loss vs. clock frequency.
This is only about the BBD block... the above answers mainly about R18, but notice R4 and R8 will affect BBD's Bias voltage too.

This is indeed a complex circuit... and, worse, its functioning depends on chips "discrepancies", this will force you to know how chips actually work and what chip parameters may vary, this will make "your" build different by almost every other build...
you should study each functional block, spend evenings on breadboard and circuit simulators and do some homeworks by yourself, then eventually come back with the right questions.

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

well, it's a big challenge indeed.... which makes iznogouds comments about "not worth it" understandable.
But I guess I'm to silly / stupid to give up :-)
It's just going to take time and learning - though I'll of course be very thankfull for all the nudges in the right direction I can get :-)

Anyways, first I need to get the schematic drawn up right - and I'd like someone to take a second look on something, compared to the drawings posted in the start of the thread;
In the schematic, pin 2 of the MN3007 is going to pin6 on the board connector, and pin 6 of the MN3007 is going to pin5 on the board connector.
However, when I look at the drawings, (SCF layout2.jpg) it's the other way around: MN3007-pin2 -> connector-pin5 and MN3007-pin6 -> connector-pin6.
Now, on the drawing, the chip is a TDA2107, but from what info I could find online, they should be the same chip, pin compatable and all.

Can anyone tell me if I'm seeing it wrong, or??

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

Fender3D wrote:First you must know how BBD works...
If you knew how BBD works you'd realize you can swap pins 2 and 6, since they're clock's pins....

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Fender3D wrote:
Fender3D wrote:First you must know how BBD works...
If you knew how BBD works
Thanks for the info.
I actually am reading up on BBD's... found this site which seems quite informative:
http://www.electrosmash.com/mn3007-buck ... de-devices
Of course there's a lot to learn, so it takes a little time to understand everything.
But maybe if I can have the patience to try to learn, people can have the patience to keep helping me in the right direction.

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Ok, I've been reading up some on BBD's and studying both the MN3007 datasheet, and the SCF schematic.
I'm trying my best to wrap my head around it, but I have a few questions that I hope you guys can answer.

I'm reading a book called "audio IC users handbook", which among other things describe BBD's.
I understand the bucket analogy, where the quantized signal gets passed to the next "bucket", when that bucket fills the previous one - or in other words, the "empty" part of each bucket is passed on.

When looking at the SCF schematic, I was surprised to see that the Vdd pin of the MN3007 was connected to ground, and the Gnd pin was connected to 15V. Looking at the datasheet I noticed that it actually does list a suggestion of minus 15 volts for Vdd, so all good then. Though I don't fully understand the reason for not using plus 15 volts.

What I'm struggling most with, is getting a full understanding of Vgg.
From what I read, Vgg is used to bias some of the transistors. Am I understanding it correctly, if I say this means that setting the voltage so that those transistors tranfers the signal at unity gain??
So Vgg should stay at the same voltage constantly? (which seems to harmonize with the -14 volts suggested in the datasheet). I also read somewhere, that it was a good practice to smooth the Vgg supply.
But I'm having trouble understanding what is up with the resistor-cap-diode arrangement (R17,C14,D4-5) between the Vgg pin and one of the clock pins (5).

Could anyone enlighten me on these points?

Regards
Anders

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

lol
my last 4 tips on my above post...

BBDs signal loss increases when clock frequency raises.
You can mitigate that loss by trimming Vgg.
Since the loss is quite heavy, TC chose to trim Vgg while raising input signal (by IC3b).
The easiest way, was to rectify clock thus getting a DC voltage following clock frequency.

MN3007 needs a negative voltage @ Vdd's pin respect GND's pin.
That's what happens there...

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Fender3D:

First:
Thanks for your tips - I'll try to wrap my head around them.

Second:
Your first remark sounds like you wont give any more tips/hints or answer any more questions of mine.
That's actually ok - you have no obligation to do so!
But, forgive me if I'm wrong, it also sounds like you think it's wrong of me to ask the questions I do.
I'm coming to this forum and asking my questions to learn - not to annoy.
Of course I have to do my "homework" by reading up on things myself, but eventually - like pretty much everybody else - I read things that I don't completely understand, or that simply doesn't make sense to me. Instead of just giving up, this is where I try to ask questions - that actually is one of the ways learning works...
It might not be the "right" questions that you feel like answering, but they are likely the questions that might help me get past my current stumbling block.
Again, you don't have to answer any question of mine - you can just.... not answer. My questions aren't for you alone - they are for anyone who'd be willing to help me.
Just please respect that I'm doing my best with the information that I can find.

And again: Thankyou a lot for your answers to my questions so far - you actually have helped me a lot! :D

/Anders

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

Anders,

I'm sorry if I gave that impression, english is not my mother tongue, then, when I reply, I tend to be as concise as I can;
you're not annoying at all, I'm glad to give help whenever I can.

And I know what you're thinking, since I had to learn once, too (and still I have to).
I'm not looking for the "best" or "rightest" question, I would like a question coming from your experience with the circuit.
Especially with such a tough schematic like this, with issues lurking as I said before.

I mean, i.e. if you try breadboarding it, you'll have other questions to ask, focused on thread and, important, questions which everyone building this, will likely ask.
Playing with the real thing, is way more useful than reading pages and comments on the net.

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Ok, sorry for my misunderstanding - and again, thanks a LOT for your help this far! :)

Actually I'm trying to follow what you wrote some posts above: First I need to understand the BBD chip... this is still what I'm working on.
Just your recent tips shows that TC made some decisions on the Vgg supply, that I didn't understand, even though I'm starting to get a handle of the working of the BBD chip itself.
I'll bet that there are plenty of places in the circuit, where they made such decisions based on a very high technical knowledge, that let them think a bit out of the box and go: "Hey, what if we took the thingamabob and whacked it around with a dead chicken... ".
Actually I'm still a little uncertain about the whole Vgg supply... The information I've found so far are a bit vague on this topic, and I haven't been able to find other circuit schematics that show a similar setup.
I am slowly starting to breadboard the circuit as best I can - I got the chips for it in today :D
I'll probably try to reduce the circuit to just the chorus part though - this is what I'm mainly interested in, and it could help to make it slightly less complex.

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

arjepsen wrote:...I'll probably try to reduce the circuit to just the chorus part....
It will be easier if you draw 3 different schematics, with the switches engaged for the 3 modes.
Follow the switch table on 1st page top left

S5 just smooths LFO.

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

I'm almost done with the breadboard layout.
Now I'm getting some funky problems with powering it.
At first I used a 2x15V trafo (2 x 1,3 VA) - and paralleled the secondaries. Rectifier bridge -> 220µF cap -> TS7815 -> 100µF

Unfortunately I was unable to get the supply voltage to the 7815 high enough for it to deliver 15V

Now I'm trying a Talema 70034 (2x18V - 2 x 3,5 VA) with paralleled secondaries. It still doesnt get the voltages high enough.
(right now it's 13,5 volts after the bridge, and 11,5 after the 7815.
The trafo and regulator both gets warm after a short while.
I thought this trafo had juice enough for the circuit... am I wrong here?

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Well, I started trying to find what uses all the power, so I lifted the power for the various parts of the circuit.
Turns out the NE570 was the culprit. Out of curiosity I put a 1R resistor from pin 4 to ground, and measuered how many milliamps was pulled through it...
...
ehm.,..
...
just about 560 milliamps......

:shock:

What on earth could posses it to draw that much?
I'be been going over the layout, but I just can't find anything wrong, according to the schematic.
Could the trimmer settings or the choice of that R4 resistor make it do this?

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

R4 sets the DC bias at the compressor output.
it is important since this same bias voltage is the same for the following chips, especially BBD.
but it won't make it draw 0.5A...

If no layout errors, check for auto oscillation, provided the chip is ok...

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

ok, my first two ne570 are fried. I have a pt80 delay that I checked them out in, and they give a nasty hum, and there's no delay.
Before I use my last ne570 I really want to try to figure out why they died.

Would the chip be damaged by supplying higher than 3 volts at the trim pin 8 and 9?

I'm gleaning over the datasheets and other literature I can find, trying to get better aquinted with this chip.
In the SCF there's a couple of diodes (D2/D3) to a 3k6 to gnd and 13k to pin 7. Would this mean that the tops/bottom of the input signal gets sent to the output pin?
In the datasheets the circuits seems to consistantly use pin 6 for input. As far as I can tell, the only difference to using pin 5 is a 20k resistor in series with the invering input of the integrated opamp.
But the SCF seems to use pin 5 for input, thus not having any resistor in series with the input. Shouldn't there be?

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

TCE relied on external parts to set desired gain and output bias voltage, I think they avoided using 570's R3.
R106, 108 and 110 act as input gain resistor(s).

At pins 8 and 9 you may safely supply 3.6V (chip's internal bias voltage), with 330K series resistor(s) I guess you may raise this voltage

D2 and D3 work in a tube screamer way to clip higher input signals.

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Fender3D wrote: At pins 8 and 9 you may safely supply 3.6V (chip's internal bias voltage), with 330K series resistor(s) I guess you may raise this voltage
Yeah, that's what I was guessing at, from what I could find in datasheets etc. But could I have damaged the chip by supplying above 3.6v?
Fender3D wrote: D2 and D3 work in a tube screamer way to clip higher input signals.
[/quote]
Yeah, I get that the diodes clip the signal, but I'm not sure why they would do that. Doesn't that distort the signal?
But instead of clipping to ground, it seems to be sent to the output pin 7, so am I understanding correctly, if I presume that the clipped part gets added back the signal path?

Yet another thing:
I've tried to do only the chorus part, so I've left out the switching chip (IC103).
As far as I can tell, this means that the circuit would act as though switch S1 and S3 are conducting. So I've left out R106, and just made a connection where S1 and S3 would be. Also S4 just shorts out R101, so I've just left R101 in place.
Is this right?
I also wonder if leaving that chip out could mess with the circuit in other ways?

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Good news.
I had one NE570 chip left. This one was slightly different from the others, and has "philips" written on it.
I was a bit unhappy about trying out my last chip that was known to work, but anyways....

I disconnected the trim controls for pin 8/9 and put a 220pf to ground from both, as per the datasheet.
After taking a deep breath, I connected power, and checked for sound at various points.
I could now confirm that sound was being passed on to the BBD - so far so good. However, no output.... hmmm..
I did some scoping on the clock pins, and it turned out that the clock didn't ...ehm.... "clock". (cluck?) :lol:
I found a few wrong connections, and fired it up again.
Woohooo - the circuit now... ehm.... "choruses"..... :lol:

Now I can finaly begin to do the tuning of the whole circuit.

I read in a book though, that the NE570 opamp isn't that good, and that using an external one might give a better sound. Wonder if it would have any positive effect on this circuit?

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Fender3D wrote: VR4 trim for lowest THD @ IC2's pin 7
Ok, I'm trying to read up on how to measure THD - and it seems like it's a bit complicated, and requires special equipment.
But most of what I read also seems to be aimed at hifi amps.
Is there an easy way to do this trimming in this circuit? Can it be done with a regular digital scope?

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

arjepsen wrote:Is there an easy way to do this trimming in this circuit? Can it be done with a regular digital scope?
NE's THD is mainly 2nd harmonic, and you get it (THD) when you hard trigger the gain cell.
the best method is feeding signal with a tone burst generator and adjust for the minimum bounce with tone bursts. (you have a straight line (DC) amidst bursts, adjust when this line is perfectly straight).
Some o-scopes measure THD and should "see" it

You may approximate this setting by measuring DC voltage @ pin 7 both with no signal and with a rather high signal, then trim for the same value at both conditions

BTW
this is a really picky setting in "guitar world", you might as well do the same Boss, Ibanez and tons of other builders did:
toss the trimmers and stick a small capacitor @ pin 8 and 9

User avatar
arjepsen
Information
Posts: 24
Joined: 18 Apr 2010, 14:12
Has thanked: 5 times

Post by arjepsen »

Thanks for your guide, I'll try that.

I actually came across some software that should be able to work as both a scope and spectrum analyser using a good quality soundcard: Visual Analyser
http://www.sillanumsoft.org/

From looking at the program, it seems it might be able to measure THD.
I've been trying to use it, and managed to make it output a 1khz sine wave.
But first I'm a little unsure about where to input this signal - and also I had some trouble with the amplitude being too high, so it overloaded the circuit.
Anyways, I was trying to input the wave before C5, and wanted to then measure at pin7 of the 570. But I'm not sure if that will work.... from doing some audio probing, I can't detect any signal before C5. I can get a signal at the wiper of the intensity pot, but the signal at pin 7 is louder.... I presume I'd need the signal to have the same amplitude to measure THD?
Am I way off here, or am I going in the right direction?

Post Reply